Objective To test the hypothesis that hospital referral for unexplained abdominal pain might occur less frequently in oral contraceptive (OC) users and more frequently in intrauterine device (IUD) users than in other women.

Design Prospective cohort study of 17 032 women using different methods of contraception [the Oxford-Family Planning Association (FPA) contraceptive study].

Outcome measure Referral to hospital for unexplained abdominal pain (including pelvic pain) coded to ICD rubric 785.5 in the 8th Revision of the ICD.

Results Referral was decreased by 13% (95% CI 0%-25%) in current or recent users of OCs in comparison with non-users, a difference that just reached statistical significance. No significant difference was found in the corresponding analysis for IUD users. Past users of OCs had much the same risk of referral as non-users, but the risk of referral in past users of an IUD was increased at all intervals examined, varying from 17% (95% CI -14%-55%) 13-24 months after discontinuation to 56% (95% CI 15%-107%) 73-96 months after discontinuation. Referral was 29% (95% CI 11%-50%) higher in smokers of 15 or more cigarettes a day than in non-smokers.

Conclusion Referral to hospital for unexplained abdominal pain may be slightly reduced in current or recent OC users and moderately increased in past IUD users. The types of IUD used most commonly in the Oxford-FPA study are, however, no longer in current use today. Hospital referral for unexplained abdominal pain seems to be moderately increased in heavy cigarette smokers.